LLM Data Annotation & Text Classification Specialist
I worked on a large-scale AI training project focused on improving Large Language Model (LLM) performance through high-quality data annotation and evaluation. My responsibilities included labeling and classifying text data, performing Named Entity Recognition (NER), rating AI-generated responses based on accuracy, relevance, and safety, and writing structured prompt-response pairs for supervised fine-tuning (SFT). I also conducted RLHF-based evaluations to enhance model alignment and reduce bias. The project involved annotating over 15,000+ text samples across multiple domains including customer service, healthcare, and general knowledge. I strictly followed annotation guidelines to ensure consistency, maintained high inter-annotator agreement scores, and adhered to data privacy and confidentiality standards. Quality assurance processes included peer reviews, multi-stage validation, and feedback loops to continuously improve annotation accuracy and model output performance.